Constrained graph-based semi-supervised learning with higher order regularization

Constrained graph-based semi-supervised learning with higher order regularization

Graph-based semi-supervised learning (SSL) algorithms have been widely studied in the last few years. Most of these algorithms were designed from unconstrained optimization problems using a Laplacian regularizer term as smoothness functional in an attempt to reflect the intrinsic geometric struc...

Full description

Bibliographic Details
Main Author: Celso Andre Rodrigues de Sousa
Other Authors: Gustavo Enrique de Almeida Prado Alves Batista
Language:English
Published: Universidade de São Paulo 2017
Subjects:
Aprendizado semissupervisionado
Métodos baseados em grafos
Otimização restrita
Regularização de ordem elevada
Constrained optimization
Graph-based methods
Higher order regularization
Semi-supervised learning
Online Access:http://www.teses.usp.br/teses/disponiveis/55/55134/tde-08122017-102557/

Internet

http://www.teses.usp.br/teses/disponiveis/55/55134/tde-08122017-102557/

Similar Items